CHAPTER XIII THE THEORY OF INSURANCE AS APPLIED TO EMPLOYERS' LIABILITY AND WORKMEN'S COMPENSATION. Insurance, from the viewpoint of most people, is an institution which provides an opportunity for securing themselves against financial loss of many kinds by the payment of a stated annual sum proportioned to the extent of the security. To secure themselves against loss by fire they pay fire insurance companies an annual premium, in consideration of which the companies will indemnify them for any damage by fire to the property insured, not exceeding the amount stated in the policy. To secure their families against loss of their income through death they make similar payments to a life insurance company. Practically every variety of financial loss may be prepared for in this manner. When the policyholder pays his premium to the insuring company and receives in return a guarantee against loss to the extent of the sum named in the policy, he has relieved himself of a risk and has brought certainty into his affairs where before uncertainty existed. But it might seem that the insurance company, which has taken over the risk, has placed itself in a more uncertain position by adding to the possible losses which it may be called upon to indemnify a service for which the premium might appear disproportionately small. As a matter of fact the addition of a new risk, with the payment of a scientifically calculated premium, increases the certainty with which the affairs of the company may be conducted, though not in so simple a fashion as is the case with the individual who pays the premium. The insurance company is able to assume additional risks with increasing certainty by reason of the application in its business of the theory of probability as applied to large groups of risks. To comprehend the insurance business in its fundamentals it is necessary to understand this theory on which it is founded. THE THEORY OF PROBABILITY The use of the theory of probability is based on the reasoned conviction that a knowledge of the past is a sufficient guide to events of the future; that, given the same conditions, we may expect the same results. According to this doctrine, if one desires to predict the results of certain present conditions, it is only necessary to learn what results have already been produced by exactly similar conditions. But exactly corresponding conditions are difficult or impossible to find and, were they essential to the operation of the theory, little practical use could be made of it. It is possible, however, by the accumulation of a large number of cases, to secure conditions which, as a whole, approximate those of the past and whose results may be expected to correspond with past results. Accuracy of the Theory.-The accuracy of any prediction based on the theory of probabilities, or the relative approximation of theoretical and actual experience, depends on three factors. (1) the degree of correspondence between the two sets of conditions, (2) the accuracy of the data, and (3) the number of cases considered in each set. The use of industrial accident statistics to determine probable future accident rates will serve as an example of the influence of the first two factors. Suppose that the accident rate for the past three years in cotton factories averaged sixty-five per one thousand employees per year. Suppose further that during the coming year many new devices for the prevention of accidents are installed, and that more careful attention is given to the reporting of accidents. Knowing these facts, may we consider the average of the past three years a reliable index to the number of accidents during the coming year? Evidently not, for the conditions given differ; in the past there were not the same facilities for the prevention of accidents, and figures based on past conditions could be used only to indicate a maximum rate which the better equipped factories will probably not attain. But there is also an improvement in methods of accident reporting which suggests that the statistics of the last three years were probably inaccurate. More careful reporting may bring in accounts of accidents which before would have passed unnoticed and the apparent increase from this source may offset the tendency of the rate to decrease as a result of safety work. For statistics to be of the greatest service for the prediction of future |